Weekday/Workday for EU locales

classic Classic list List threaded Threaded
18 messages Options
Reply | Threaded
Open this post in threaded view
|

Weekday/Workday for EU locales

Jeff Law

There are a multitude of bugs in the database related to the wrong first
weekday/workday for locales within the EU.

In the Official Journal of the European Union:

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:C:2004:78E:0447:0447:EN:PDF

(2004/C 78 E/0469)
by Christopher Huhne (ELDR) to the Commission
(14 October 2003)
Subject: Standardisation of computer dates
Will the Commission state whether there is any EU legislation that sets
a standard format for dates in
computer software?
Answer given by Mr Liikanen on behalf of the Commission
(24 November 2003)
The format for dates in computer software has not been regulated by
Community legislation.
However, in 1992 the European Committee For Standardization (1) has
adopted ISO 8601 under the
standard EN 28601 to end the traditional confusion involving periods,
slashes, the order of the numbers,
and other date formats formerly used throughout Europe.

So it seems ISO 8601 is the standard we need to refer to to for such
information within the EU.

8601:

http://dotat.at/tmp/ISO_8601-2004_E.pdf

Section 2.2.8, calendar week:

time interval of seven calendar days starting with a Monday


So can we, pretty please, agree that at least for EU countries that
fixing the first weekday and first workday to Monday within the
appropriate locales is the right thing to do?

jeff
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales

Carlos O'Donell-2
On Fri, Apr 13, 2012 at 3:08 PM, Jeff Law <[hidden email]> wrote:
> There are a multitude of bugs in the database related to the wrong first
> weekday/workday for locales within the EU.

Could you please file bugzilla issues for each locale so we can fix them?

> So can we, pretty please, agree that at least for EU countries that fixing
> the first weekday and first workday to Monday within the appropriate locales
> is the right thing to do?

I have reviewed ISO 8601. Your presented evidence is certainly very
convincing. If I had to review a patch to fix such an issue I would
agree with you that first weekday and workday should be Monday (within
the appropriate locales).

Next steps:

(a) Gather distro maintainer consensus by sending an email to
libc-alpha, CC all the distro maintainers asking for comment, give
them two weeks to respond.

http://sourceware.org/glibc/wiki/MAINTAINERS#Distribution_Maintainers

(b) File individual bugs for locales which are wrong according to ISO 8601.

(c) If the distros respond favourably, start submitting patches to fix
all the bugs, which will be accepted given the presented evidence.

Comments?

Cheers,
Carlos.
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales

Andreas Jaeger-8
On 04/19/2012 06:45 PM, Carlos O'Donell wrote:
> On Fri, Apr 13, 2012 at 3:08 PM, Jeff Law<[hidden email]>  wrote:
>> There are a multitude of bugs in the database related to the wrong first
>> weekday/workday for locales within the EU.
>
> Could you please file bugzilla issues for each locale so we can fix them?

Shouldn't one mentioning all enough - especially if it contains a patch? ;)

>> So can we, pretty please, agree that at least for EU countries that fixing
>> the first weekday and first workday to Monday within the appropriate locales
>> is the right thing to do?
>
> I have reviewed ISO 8601. Your presented evidence is certainly very
> convincing. If I had to review a patch to fix such an issue I would
> agree with you that first weekday and workday should be Monday (within
> the appropriate locales).
>
> Next steps:
>
> (a) Gather distro maintainer consensus by sending an email to
> libc-alpha, CC all the distro maintainers asking for comment, give
> them two weeks to respond.

I'd like to have a list of wrong locales first.

> http://sourceware.org/glibc/wiki/MAINTAINERS#Distribution_Maintainers
>
> (b) File individual bugs for locales which are wrong according to ISO 8601.
>
> (c) If the distros respond favourably, start submitting patches to fix
> all the bugs, which will be accepted given the presented evidence.
>
> Comments?

Andreas
--
  Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi
   SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
    GF: Jeff Hawn,Jennifer Guild,Felix Imendörffer,HRB16746 (AG Nürnberg)
     GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales

Petr Baudis
On Fri, Apr 20, 2012 at 10:32:26AM +0200, Andreas Jaeger wrote:
> On 04/19/2012 06:45 PM, Carlos O'Donell wrote:
> >On Fri, Apr 13, 2012 at 3:08 PM, Jeff Law<[hidden email]>  wrote:
> >>There are a multitude of bugs in the database related to the wrong first
> >>weekday/workday for locales within the EU.
> >
> >Could you please file bugzilla issues for each locale so we can fix them?
>
> Shouldn't one mentioning all enough - especially if it contains a patch? ;)

I agree, bugzilla issues for each locale seems like a needless
bureaucracy here. Having this in a single issue would carry less noise.

> >>So can we, pretty please, agree that at least for EU countries that fixing
> >>the first weekday and first workday to Monday within the appropriate locales
> >>is the right thing to do?
> >
> >I have reviewed ISO 8601. Your presented evidence is certainly very
> >convincing. If I had to review a patch to fix such an issue I would
> >agree with you that first weekday and workday should be Monday (within
> >the appropriate locales).
> >
> >Next steps:
> >
> >(a) Gather distro maintainer consensus by sending an email to
> >libc-alpha, CC all the distro maintainers asking for comment, give
> >them two weeks to respond.
>
> I'd like to have a list of wrong locales first.
>
> >http://sourceware.org/glibc/wiki/MAINTAINERS#Distribution_Maintainers

It would be good to also try sending notification mails to
contact emails of all the concerned locales, in case any locale
maintainer has any reservations. When you have a list of locales,
doing this should be fairly trivial.

--
                                Petr "Pasky" Baudis
        Smart data structures and dumb code works a lot better
        than the other way around.  -- Eric S. Raymond
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales

Jeff Law
On 04/20/2012 05:02 AM, Petr Baudis wrote:

> On Fri, Apr 20, 2012 at 10:32:26AM +0200, Andreas Jaeger wrote:
>> On 04/19/2012 06:45 PM, Carlos O'Donell wrote:
>>> On Fri, Apr 13, 2012 at 3:08 PM, Jeff Law<[hidden email]>   wrote:
>>>> There are a multitude of bugs in the database related to the wrong first
>>>> weekday/workday for locales within the EU.
>>>
>>> Could you please file bugzilla issues for each locale so we can fix them?
>>
>> Shouldn't one mentioning all enough - especially if it contains a patch? ;)
>
> I agree, bugzilla issues for each locale seems like a needless
> bureaucracy here. Having this in a single issue would carry less noise.
Agreed.  One BZ for the overarching issue, the existing locale-specific
BZs around this issue would then be closed as duplicates.  The last
thing I want to do is "make work".

Creating a BZ for each locale that needs fixing would probably take
longer than fixing the bug itself across all the locales..

>>>
>>> (a) Gather distro maintainer consensus by sending an email to
>>> libc-alpha, CC all the distro maintainers asking for comment, give
>>> them two weeks to respond.
>>
>> I'd like to have a list of wrong locales first.
>>
>>> http://sourceware.org/glibc/wiki/MAINTAINERS#Distribution_Maintainers
>
> It would be good to also try sending notification mails to
> contact emails of all the concerned locales, in case any locale
> maintainer has any reservations. When you have a list of locales,
> doing this should be fairly trivial.
I'll pull together the list of locales and send the message.


jeff

Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Jeff Law
In reply to this post by Carlos O'Donell-2

There's a long standing problem with handling of the first weekday and
first workday in glibc's various EU locales.  Carlos & I have both
reviewed ISO 8601 and our reading is that for all EU members Monday is
the first weekday and first workday.

I'd like to get this addressed in all the EU locales, preferably
en-masse rather than continuing to fault in a fix for each one when
someone complains.

The following locales are broken in this regard:
de_BE, de_BE@euro, nl_BE, nl_BE@euro, wa_BE, wa_BE@euro, el_CY,  tr_CY,
br_FR, br_FR@euro, oc_FR, el_GR, el_GR@euro, en_IE, en_IE@euro, ga_IE,
ga_IE@euro, it_IT, it_IT@euro, lij_IT, lv_LV, lt_LT, de_LU, de_LU@euro,
fr_LU, fr_LU@euro, mt_MT, fy_NL, pt_PT, pt_PT@euro, an_ES, eu_ES,
eu_ES@euro, gl_ES, gs_ES@euro, cy_GB, gd_GB, gv_GB, kw_GB,

ast_ES just needs the first_workday fix.

Unless objections are noted, the plan is to fix these locals on June 6,
2012.  If you're objecting to a fix, please state specifically which
locale(s) you believe should not be changed and why.


For reference, the following locals are OK:
de_AT, de_AT@euro, fr_BE, fe_BE@euro, li_BE,  be_BG, cs_CZ, da_DK,
en_DK, et_EE, fi_FI, fi_FI@euro, sv_FI, sv_FI@euro, ca_FR, fr_FR,
fr_FR@euro, de_DE, de_DE@euro, fy_DE, hsb_DE, nds_DE, hu_HU,  ca_IT,
fur_IT, sc_IT, lb_LU, li_NL, nds_NL, nl_NL, nl_NL@euro, csb_PL, pl_PL,
ro_RO, sl_SI, ca_ES, ca_ES@euro, es_ES, es_ES@euro, sv_SE, en_GB, sk_SK


Jeff
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Andreas Jaeger-8
On 05/23/2012 07:57 PM, Jeff Law wrote:

>
> There's a long standing problem with handling of the first weekday and
> first workday in glibc's various EU locales. Carlos & I have both
> reviewed ISO 8601 and our reading is that for all EU members Monday is
> the first weekday and first workday.
>
> I'd like to get this addressed in all the EU locales, preferably
> en-masse rather than continuing to fault in a fix for each one when
> someone complains.
>
> The following locales are broken in this regard:
> de_BE, de_BE@euro, nl_BE, nl_BE@euro, wa_BE, wa_BE@euro, el_CY, tr_CY,
> br_FR, br_FR@euro, oc_FR, el_GR, el_GR@euro, en_IE, en_IE@euro, ga_IE,
> ga_IE@euro, it_IT, it_IT@euro, lij_IT, lv_LV, lt_LT, de_LU, de_LU@euro,
> fr_LU, fr_LU@euro, mt_MT, fy_NL, pt_PT, pt_PT@euro, an_ES, eu_ES,
> eu_ES@euro, gl_ES, gs_ES@euro, cy_GB, gd_GB, gv_GB, kw_GB,
>
> ast_ES just needs the first_workday fix.

There's also http://sourceware.org/bugzilla/show_bug.cgi?id=13223 
speaking about tr_TR.

Not EU - but same problem.

> Unless objections are noted, the plan is to fix these locals on June 6,
> 2012. If you're objecting to a fix, please state specifically which
> locale(s) you believe should not be changed and why.
>
>
> For reference, the following locals are OK:
> de_AT, de_AT@euro, fr_BE, fe_BE@euro, li_BE, be_BG, cs_CZ, da_DK, en_DK,
> et_EE, fi_FI, fi_FI@euro, sv_FI, sv_FI@euro, ca_FR, fr_FR, fr_FR@euro,
> de_DE, de_DE@euro, fy_DE, hsb_DE, nds_DE, hu_HU, ca_IT, fur_IT, sc_IT,
> lb_LU, li_NL, nds_NL, nl_NL, nl_NL@euro, csb_PL, pl_PL, ro_RO, sl_SI,
> ca_ES, ca_ES@euro, es_ES, es_ES@euro, sv_SE, en_GB, sk_SK

no objections from me,
Andreas
--
  Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi
   SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
    GF: Jeff Hawn,Jennifer Guild,Felix Imendörffer,HRB16746 (AG Nürnberg)
     GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Michael Matz
In reply to this post by Jeff Law
Hi,

On Wed, 23 May 2012, Jeff Law wrote:

> There's a long standing problem with handling of the first weekday and
> first workday in glibc's various EU locales.  Carlos & I have both
> reviewed ISO 8601 and our reading is that for all EU members Monday is
> the first weekday and first workday.

ISO 8601 makes no provisions for changing the first weekday according to
locale; it is Monday for whomever chooses to follow that standard in its
entirety.  Most (all?) european countries happen to follow it in this
aspect :)

> I'd like to get this addressed in all the EU locales, preferably
> en-masse rather than continuing to fault in a fix for each one when
> someone complains.

Shouldn't localedata/i18n (although it's not used as template for any
LC_TIME) and programs/ld-time.c then simply be changed to use the ISO 8601
defaults?  Because, of all the 377 (on my system here) locales that
currently use Sunday as first weekday most of them have it only because of
these defaults.  Only en_US, es_CR, et_EE, ks_IN@devanagari, shs_CA and
wa_BE set it explicitely, and of those only wa_BE is used as template for
a further locale's LC_TIME (wa_BE@euro).  That is, 370 locales have first
weekday=1 just because of the default in programs/ld-time.c.  Nobody can
tell me that this was a concious decision for each one of these locales,
rather than just accident.


Ciao,
Michael.
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Jeff Law
On 05/24/2012 07:16 AM, Michael Matz wrote:
> ISO 8601 makes no provisions for changing the first weekday according to
> locale; it is Monday for whomever chooses to follow that standard in its
> entirety.  Most (all?) european countries happen to follow it in this
> aspect :)
Well, the doc requirements in the past to get this kind of thing fixed
have been, umm, quite burdensome.  Fixing it for the EU seems like a
no-brainer given the commission's clarification that they're following
ISO 8601.

For countries outside the EU, I haven't done the research necessary to
meet the doc requirements we've had for locale fixes.


Jeff
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Petr Baudis
In reply to this post by Jeff Law
  Hi!

On Wed, May 23, 2012 at 11:57:17AM -0600, Jeff Law wrote:
> Unless objections are noted, the plan is to fix these locals on June
> 6, 2012.  If you're objecting to a fix, please state specifically
> which locale(s) you believe should not be changed and why.

  Excellent, thanks for doing this!

  However, the trunk freeze is planned on June 4, isn't it? So it should
land in trunk sooner (ideally leave few days buffer to allow for delays,
I'd say).

                                Petr "Pasky" Baudis
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Jeff Law
On 05/24/2012 02:03 PM, Petr Baudis wrote:

>    Hi!
>
> On Wed, May 23, 2012 at 11:57:17AM -0600, Jeff Law wrote:
>> Unless objections are noted, the plan is to fix these locals on June
>> 6, 2012.  If you're objecting to a fix, please state specifically
>> which locale(s) you believe should not be changed and why.
>
>    Excellent, thanks for doing this!
>
>    However, the trunk freeze is planned on June 4, isn't it? So it should
> land in trunk sooner (ideally leave few days buffer to allow for delays,
> I'd say).
Carlos wanted a 2wk warning period.  When combined with a short vacation
I have planned at the start of June, well, you end up with a June 6 date.

I can certainly post the patch to fix the EU locales prior to
disappearing and let Carlos or someone else run with it on the 4th or
just prior to the 4th if they want.
jeff
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Michael Matz
In reply to this post by Jeff Law
Hi,

On Thu, 24 May 2012, Jeff Law wrote:

> Well, the doc requirements in the past to get this kind of thing fixed have
> been, umm, quite burdensome.  Fixing it for the EU seems like a no-brainer
> given the commission's clarification that they're following ISO 8601.
>
> For countries outside the EU, I haven't done the research necessary to meet
> the doc requirements we've had for locale fixes.

I understand.  There are even fewer useful documents about weekday
conventions in legislations than there are about time changes.  But OTOH
the same argument can be used against the initial and still current state:
"who documented and where are the so produced documents that defined the
current state to be correct?".  Just because Uli required elaborate proofs
for _changes_ of the current status doesn't in any case imply the even
remote correctness of that state.  In fact I think it was completely
arbitrary and in most cases just wrong (exactly because he didn't count
hear-say evidence like "nobody in our country considers Sunday first day";
meaning he took lack of official proof for incorrectness to be proof of
correctness of current state; which of course is absurd).

Are we even discussing these topics on the right list?  Half of the mails
in CC bounced, *-locale(s?) weren't added, and IMO, from reading the
lists, the timezone list (olson) might have the diversest (if that's a
word) set of persons that could usefully contribute to the topic at hand;
even though it's not directly timezone related that list has the most
persons on it that are remotely interested (and able to produce links to
official documents) in time/date topics.

That said, I of course am completely for changing the currently broken
locales.  I just doubt that if one ventures into the other locales that
one would find those are correct regarding this topic.  So fixing the
european locales only is useful, but I think ultimately hypocritical.


Ciao,
Michael.
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Jeff Law
On 05/24/2012 04:53 PM, Michael Matz wrote:

>
> I understand.  There are even fewer useful documents about weekday
> conventions in legislations than there are about time changes.  But OTOH
> the same argument can be used against the initial and still current state:
> "who documented and where are the so produced documents that defined the
> current state to be correct?".  Just because Uli required elaborate proofs
> for _changes_ of the current status doesn't in any case imply the even
> remote correctness of that state.  In fact I think it was completely
> arbitrary and in most cases just wrong (exactly because he didn't count
> hear-say evidence like "nobody in our country considers Sunday first day";
> meaning he took lack of official proof for incorrectness to be proof of
> correctness of current state; which of course is absurd).
I certainly understand your arguments and I certainly found the
situation, umm, inconsistent.  A bit of web searching will show the lack
of locale fixes in glibc as being a source of significant frustration.

Regardless, I wasn't ready to try and fundamentally change how we do
things, but instead make a significant improvement in the state of the
locales.  ISTM tackling these other issues is independent of moving
forward on cases where we can meet the burden of proof that has
traditionally been required.

>
> Are we even discussing these topics on the right list?  Half of the mails
> in CC bounced,
Vast majority bounced.  I think even the one gent who was responsible
for most of the locales bounced :(



> *-locale(s?) weren't added,
-locales has been a complete black hole based on my experience.  Adding
it would have been pointless IMHO.


  and IMO, from reading the
> lists, the timezone list (olson) might have the diversest (if that's a
> word) set of persons that could usefully contribute to the topic at hand;
> even though it's not directly timezone related that list has the most
> persons on it that are remotely interested (and able to produce links to
> official documents) in time/date topics.
I think we could certainly bring them into the larger discussion.  Red
Hat has folks internally that deal with tz nonsense all the time as well
(Petr M).

>
> That said, I of course am completely for changing the currently broken
> locales.  I just doubt that if one ventures into the other locales that
> one would find those are correct regarding this topic.  So fixing the
> european locales only is useful, but I think ultimately hypocritical.
I'm certainly not suggesting we stop with fixing the EU locales, it's
step #1 and IMHO a step we ought to take now rather than wait on a new
set of policies & procedures and potential input from new contributors
such as olson, Petr M., etc.

jeff

Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Carlos O'Donell-4
On 5/30/2012 1:36 AM, Jeff Law wrote:
>> Are we even discussing these topics on the right list?  Half of the
>> mails in CC bounced,
> Vast majority bounced.  I think even the one gent who was responsible
> for most of the locales bounced :(

Would you please be so kind as to checkin patches that remove emails
that bounce and set them to [hidden email]? This is the
kind of obvious patch that you can checkin without review.
 
>> *-locale(s?) weren't added,
> -locales has been a complete black hole based on my experience.
> Adding it would have been pointless IMHO.

I agree. In fact I would immediately accept a patch that scrubs
all *locale* email references to [hidden email].

We need to bring all the discussions back to libc-alpha and
start from the ground up again.

>> That said, I of course am completely for changing the currently
>> broken locales.  I just doubt that if one ventures into the other
>> locales that one would find those are correct regarding this topic.
>> So fixing the european locales only is useful, but I think
>> ultimately hypocritical.
> I'm certainly not suggesting we stop with fixing the EU locales, it's
> step #1 and IMHO a step we ought to take now rather than wait on a
> new set of policies & procedures and potential input from new
> contributors such as olson, Petr M., etc.

Thanks for driving this forward Jeff.

Cheers,
Carlos.
--
Carlos O'Donell
Mentor Graphics / CodeSourcery
[hidden email]
[hidden email]
+1 (613) 963 1026
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Jeff Law
On 05/30/2012 07:30 AM, Carlos O'Donell wrote:
> On 5/30/2012 1:36 AM, Jeff Law wrote:
>>> Are we even discussing these topics on the right list?  Half of the
>>> mails in CC bounced,
>> Vast majority bounced.  I think even the one gent who was responsible
>> for most of the locales bounced :(
>
> Would you please be so kind as to checkin patches that remove emails
> that bounce and set them to [hidden email]? This is the
> kind of obvious patch that you can checkin without review.
Do you want this to wait until after 2.16 is released?  I've got a patch
which fixes the bug contact address and starts the removal of
maintainers which are no longer reachable (just their email/phone #,
leaving their names for historical reference).

Jeff
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Carlos O'Donell-4
On 6/18/2012 4:20 PM, Jeff Law wrote:

> On 05/30/2012 07:30 AM, Carlos O'Donell wrote:
>> On 5/30/2012 1:36 AM, Jeff Law wrote:
>>>> Are we even discussing these topics on the right list?  Half of
>>>> the mails in CC bounced,
>>> Vast majority bounced.  I think even the one gent who was
>>> responsible for most of the locales bounced :(
>>
>> Would you please be so kind as to checkin patches that remove
>> emails that bounce and set them to [hidden email]? This
>> is the kind of obvious patch that you can checkin without review.
> Do you want this to wait until after 2.16 is released?  I've got a
> patch which fixes the bug contact address and starts the removal of
> maintainers which are no longer reachable (just their email/phone #,
> leaving their names for historical reference).

Lets wait until after 2.16 please.

Cheers,
Carlos.
--
Carlos O'Donell
Mentor Graphics / CodeSourcery
[hidden email]
[hidden email]
+1 (613) 963 1026
Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Jeff Law
In reply to this post by Michael Matz
On 05/24/2012 07:16 AM, Michael Matz wrote:

> Shouldn't localedata/i18n (although it's not used as template for any
> LC_TIME) and programs/ld-time.c then simply be changed to use the ISO 8601
> defaults?  Because, of all the 377 (on my system here) locales that
> currently use Sunday as first weekday most of them have it only because of
> these defaults.  Only en_US, es_CR, et_EE, ks_IN@devanagari, shs_CA and
> wa_BE set it explicitely, and of those only wa_BE is used as template for
> a further locale's LC_TIME (wa_BE@euro).  That is, 370 locales have first
> weekday=1 just because of the default in programs/ld-time.c.  Nobody can
> tell me that this was a concious decision for each one of these locales,
> rather than just accident.
So I'm happy to go forward with this, but would like to hear from the
maintainers first.

Jeff

Reply | Threaded
Open this post in threaded view
|

Re: Weekday/Workday for EU locales in glibc

Petr Baudis
On Thu, Aug 09, 2012 at 10:26:41PM -0600, Jeff Law wrote:

> On 05/24/2012 07:16 AM, Michael Matz wrote:
> >Shouldn't localedata/i18n (although it's not used as template for any
> >LC_TIME) and programs/ld-time.c then simply be changed to use the ISO 8601
> >defaults?  Because, of all the 377 (on my system here) locales that
> >currently use Sunday as first weekday most of them have it only because of
> >these defaults.  Only en_US, es_CR, et_EE, ks_IN@devanagari, shs_CA and
> >wa_BE set it explicitely, and of those only wa_BE is used as template for
> >a further locale's LC_TIME (wa_BE@euro).  That is, 370 locales have first
> >weekday=1 just because of the default in programs/ld-time.c.  Nobody can
> >tell me that this was a concious decision for each one of these locales,
> >rather than just accident.
> So I'm happy to go forward with this, but would like to hear from
> the maintainers first.

I'm feeling somewhat uneasy about changing the defaults. I would prefer
to instead explicitly set weekday in each locale, since this seems to be
a complex issue even for EU countries, could create confusion with
locale definitions that are currently floating around the net unmerged,
etc. Also, I'm not sure if that's the current state in all regards,
but it makes sense to me to have the defaults match the POSIX locale.

So let's make this an explicit action for each locale involved?

(This does not mean that it should not necessarily be en bloc change.
We might consider looking at other locale repositories and doing a simple
comparison.)

                                Petr "Pasky" Baudis